Fatores de risco para comportamento anti-social na adolescência e início da vida adulta
Data
1996
Autores
Bordin, Isabel Altenfelder Santos 
Orientadores
Mari, Jair de Jesus
Tipo
Tese de doutorado
Título da Revista
ISSN da Revista
Título de Volume
Resumo
Introdução: Violência e crime são problemas crescentes na sociedade e a
eficiência de programas preventivos depende da identificação dos fatores
de risco que atuam precocemente na vida da criança.
Objetivos: Verificar se o CAS antes dos 15 anos violento e não violento
são fatores de risco para CAS na adolescência e inicio da vida adulta na
presença de seis eventos relativos a infância, examinando a magnitude
deste efeito em ambos sexos; identificar o melhor modelo preditivo para
CAS a partir dos 15 anos nas populações masculina e feminina,
considerando tipos variados de CAS antes dos 15 anos e seis eventos
relativos i infância; verificar a influencia dos eventos relativos a
infância na persistência do CAS a partir dos 15 anos e no inicio tardio
do CAS em ambos sexos.
Desenho: Corte transversal (Ontario Health Survey Supplement).
Local: Moradores de residências particulares de áreas urbanas e rurais
da província de Ontario, Canada.
Participantes: 773 indivíduos do sexo masculino e 878 do feminino
(N=1.651) com idade entre 16-24 anos. A escolha da amostra seguiu
estágios múltiplos de estratificação. Mesmo após a perda de 10,2% dos
indivíduos com dados incompletos para a análise de regressão logística
(N=1482), o poder da amostra permaneceu adequado (>90%) para detectar
com segurança (p<0,01) até mesmo pequenos efeitos dos fatores de risco
sobre o desfecho clinico.
Medidas: Desfecho clinico definido como 3+ CAS a partir dos 15 anos.
Fatores de risco incluiram diferentes definições e combinações de CAS
violento (V) e não violento (NV) antes dos 15 anos (Tabela 1) e seis
eventos relativos A infância (abuso físico; abuso sexual; ausência de
relacionamento próximo com adultos; discórdia conjugal entre os pais;
pais com depressão, mania ou episódio psicótico e pais com CAS ou abuso
de Alcoól ou drogas). Os dados sobre todas as variáveis, exceto o abuso
na infância, foram coletados com o auxilio do CIDI modificado, um
questionário estruturado aplicado por entrevistador treinado. Os dados
sobre abuso físico e sexual foram obtidos através de questionário autoaplicado.
Análise: Cinco modelos de regressão logística foram examinados. Os
modelos número 1, 2 e 3 incluíram a amostra total, o modelo 4 incluiu os
indivíduos com CAS antes dos 15 anos e o modelo 5 incluiu os indivíduos
sem CAS antes dos 15 anos...
Rationale: Violence and crime are growing problems in our society and effectiveness of prevention programs depends on the identification of early predictors. Objectives: To examine the magnitude of effect that violent and nonviolent ASB before age 15 have on the prediction of ASB in adolescence and early adult life since turning 15 when taking into account six childhood events; to identify the best prediction model including different definitions of ASB before age 15 and six childhood events; to identify gender differences in the prediction models; to examine the influence of childhood events on persistence and late onset of ASB according to gender. Design: Cross-sectional study (Ontario Health Survey Supplement). Setting: Residents in private dwellings from urban and rural areas of Ontario, Canada. Participants: 773 males and 878 females (N=1651) aged 16-24 years. Sampling plan described as a multi-stage stratified cluster design. Only cases with complete information (N=1482) were included in the logistic regression analysis (missing=10.2%). Power (>90%) remained high enough to detect even small predictor effects on the outcome at a reasonable level of type I error (p<0,01). Measurement: Outcome defined as 3+ ASB since turning 15. Predictors included different definitions and combinations of violent (V) and nonviolent (NV) ASB before age 15 (Table 1) and six childhood variables (physical abuse; sexual abuse; absence of a confiding relationship with an adult; parent marital discord; parent depression, mania or psychotic episode; parent with ASB or alcohol-drug problems). All data except child abuse gathered using an improved version of the Composite International Diagnostic Interview (CIDI), a highly structured questionnaire administered by interviewers. Data on abuse obtained by a self-report questionnaire. Analysis: Five logistic regression models were examined. Models number 1, 2 and 3 included the entire individuals with ASB before age without ASB before age 15. Two sample, model 4 included only those 15, and model 5 included only those analysis strategies were applied to models 1, 2 and 3. First, all independent variables were forced to enter the regression equation, and second, stepwise logistic regression selected only the significant predictors (p<0.05). Stepwise logistic regression was also applied to models 4 and 5. All logistic regression analyses were performed for males and females separately...
Rationale: Violence and crime are growing problems in our society and effectiveness of prevention programs depends on the identification of early predictors. Objectives: To examine the magnitude of effect that violent and nonviolent ASB before age 15 have on the prediction of ASB in adolescence and early adult life since turning 15 when taking into account six childhood events; to identify the best prediction model including different definitions of ASB before age 15 and six childhood events; to identify gender differences in the prediction models; to examine the influence of childhood events on persistence and late onset of ASB according to gender. Design: Cross-sectional study (Ontario Health Survey Supplement). Setting: Residents in private dwellings from urban and rural areas of Ontario, Canada. Participants: 773 males and 878 females (N=1651) aged 16-24 years. Sampling plan described as a multi-stage stratified cluster design. Only cases with complete information (N=1482) were included in the logistic regression analysis (missing=10.2%). Power (>90%) remained high enough to detect even small predictor effects on the outcome at a reasonable level of type I error (p<0,01). Measurement: Outcome defined as 3+ ASB since turning 15. Predictors included different definitions and combinations of violent (V) and nonviolent (NV) ASB before age 15 (Table 1) and six childhood variables (physical abuse; sexual abuse; absence of a confiding relationship with an adult; parent marital discord; parent depression, mania or psychotic episode; parent with ASB or alcohol-drug problems). All data except child abuse gathered using an improved version of the Composite International Diagnostic Interview (CIDI), a highly structured questionnaire administered by interviewers. Data on abuse obtained by a self-report questionnaire. Analysis: Five logistic regression models were examined. Models number 1, 2 and 3 included the entire individuals with ASB before age without ASB before age 15. Two sample, model 4 included only those 15, and model 5 included only those analysis strategies were applied to models 1, 2 and 3. First, all independent variables were forced to enter the regression equation, and second, stepwise logistic regression selected only the significant predictors (p<0.05). Stepwise logistic regression was also applied to models 4 and 5. All logistic regression analyses were performed for males and females separately...
Descrição
Citação
São Paulo: [s.n.], 1996. 184 f.