Adverse childhood experiences and crime outcomes in early adulthood: A multi-method approach in a Brazilian birth cohort.

This study aimed to investigate alternative approaches to a cumulative risk score in the relationship between adverse childhood experiences (ACEs) and crime. Using data from the 1993 Pelotas (Brazil) Birth Cohort (n = 3236), we measured 12 ACEs up to 15 years, and past-year violent and non-violent crime at 22 years. We used four analytical approaches: single adversities, cumulative risk, latent class analysis, and network analysis. When examined individually, physical abuse, emotional abuse, and domestic violence were associated with both crime outcomes, whereas maternal mental illness and discrimination were associated with violent crime only, and parental divorce and poverty with non-violent crime only. There was a cumulative effect of ACEs on crime. The class with child maltreatment and household challenges was associated with both crime outcomes; exposure to household challenges and social risks was associated with violent crime only. In network models, crime showed conditional associations with physical abuse, maternal mental illness, and parental divorce. Although cumulative ACEs did associate with crime, some individual and combinations of ACEs showed particularly strong and robust effects, which were not captured by the cumulative score. Many ACEs are closely connected and/or cluster together, and the usefulness of the ACE score needs to be further evaluated.

[Child marriage in Brazil: an analysis from the National Health Survey]. / Casamento infantil no Brasil: uma análise da Pesquisa Nacional de Saúde.

This paper aims to characterize children and adolescents in a child marriage situation using data from the National Health Survey of 2013. The prevalence of child marriage (under 18 years old) was estimated according to socioeconomic, demographic, and health variables by three age categories (10-13; 14-15; 16-17 years). Prevalence ratios (PR) were estimated using Poisson regression. The prevalence of child marriage was 3.9% (n=1,168); 254 (1.8%) were under 14 years old, 285 (3.8%) were between 14 or 15 years old and 629 (8.1%) were between 16 and 17 years old. Higher probability of the outcome was observed among girls in the 14-15- and 16-17-years age groups compared to male, and those who did not attend school (observed in all age groups). Children living with four people were less likely to be in child marriage in the 14-15- and 16-17-years' age group compared to those who lived with one to two people. This study discusses the implications of the findings for preventing child marriage by focusing on gender inequalities and access to education and health services.

O objetivo deste artigo é caracterizar crianças e adolescentes que vivem em situação de casamento infantil utilizando dados da Pesquisa Nacional de Saúde de 2013. Estimou-se a prevalência de casamento infantil (de menores de 18 anos) de acordo com variáveis socioeconômicas, demográficas e de saúde, de acordo com três categorias de idade (10-13; 14-15; 16-17 anos). Razões de prevalências (RP) foram estimadas por meio de regressão de Poisson. A prevalência geral de casamento infantil foi de 3,9% (n = 1.168); 254 (1,8%) apresentavam menos de 14 anos, 285 (3,8%) tinham entre 14 ou 15 anos e 629 (8,1%) de 16 a 17 anos. A maior probabilidade do evento foi observada no sexo feminino, nas faixas etárias 14 a 15 e 16 a 17 anos. Assim como nas crianças e adolescentes que não possuíam vínculo escolar (observada em todas as faixas etárias). Já os indivíduos que moravam junto de quatro pessoas tiveram menor probabilidade de estar em um casamento infantil nas faixas etárias de 14 a 15 e 16 a 17 anos, em relação aos que moravam com uma a duas pessoas. Este estudo discute as implicações dos achados para a prevenção do casamento infantil, focando nas desigualdades de gênero e no acesso a educação e serviços de saúde.

Casamento infantil no Brasil: uma análise da Pesquisa Nacional de Saúde / Child marriage in Brazil: an analysis from the National Health Survey

Resumo O objetivo deste artigo é caracterizar crianças e adolescentes que vivem em situação de casamento infantil utilizando dados da Pesquisa Nacional de Saúde de 2013. Estimou-se a prevalência de casamento infantil (de menores de 18 anos) de acordo com variáveis socioeconômicas, demográficas e de saúde, de acordo com três categorias de idade (10-13; 14-15; 16-17 anos). Razões de prevalências (RP) foram estimadas por meio de regressão de Poisson. A prevalência geral de casamento infantil foi de 3,9% (n = 1.168); 254 (1,8%) apresentavam menos de 14 anos, 285 (3,8%) tinham entre 14 ou 15 anos e 629 (8,1%) de 16 a 17 anos. A maior probabilidade do evento foi observada no sexo feminino, nas faixas etárias 14 a 15 e 16 a 17 anos. Assim como nas crianças e adolescentes que não possuíam vínculo escolar (observada em todas as faixas etárias). Já os indivíduos que moravam junto de quatro pessoas tiveram menor probabilidade de estar em um casamento infantil nas faixas etárias de 14 a 15 e 16 a 17 anos, em relação aos que moravam com uma a duas pessoas. Este estudo discute as implicações dos achados para a prevenção do casamento infantil, focando nas desigualdades de gênero e no acesso a educação e serviços de saúde.

Abstract This paper aims to characterize children and adolescents in a child marriage situation using data from the National Health Survey of 2013. The prevalence of child marriage (under 18 years old) was estimated according to socioeconomic, demographic, and health variables by three age categories (10-13; 14-15; 16-17 years). Prevalence ratios (PR) were estimated using Poisson regression. The prevalence of child marriage was 3.9% (n=1,168); 254 (1.8%) were under 14 years old, 285 (3.8%) were between 14 or 15 years old and 629 (8.1%) were between 16 and 17 years old. Higher probability of the outcome was observed among girls in the 14-15- and 16-17-years age groups compared to male, and those who did not attend school (observed in all age groups). Children living with four people were less likely to be in child marriage in the 14-15- and 16-17-years' age group compared to those who lived with one to two people. This study discusses the implications of the findings for preventing child marriage by focusing on gender inequalities and access to education and health services.

Body mass index and mortality among community-dwelling elderly of Southern Brazil.

This study aims to measure the association between body mass index (BMI), comparing two different classifications, and mortality among community-dwelling elderly considering myopenia in Pelotas, Brazil. This is a longitudinal study started in 2014, we followed 1451 elderly people (≥ 60 years) enrolled in the "COMO VAI?" study. BMI was classified according to the World Health Organization (WHO) and the classification with specific cutoff points for older adults. Myopenia was measured by calf circumference (≤33 cm for women and ≤34 cm for men). Cox proportional-hazards models were used to test associations controlling for sociodemographic and behavioral characteristics and number of morbidities. Nearly 10% (N = 145) of the elderly died during almost three years of follow-up. We observed a L-shaped relation between BMI and mortality. Elderly with underweight had a higher mortality risk compared to those with adequate BMI in both classifications. According to the WHO classification, overweight elderly presented protection for mortality (HR: 0.58; 95% CI 0.38-0.87) when compared to those with adequate BMI. Among elderly with myopenia, overweight by WHO continued to protect against mortality, although not significantly, while those with the specific classification underweight presented a higher risk of death compared to those with normal weight (HR: 2.09; 95% CI 1.06-4.14). In conclusion the underweight increased the risk of death in community-dwelling elderly people during a follow-up of three years. The specific classification seemed to be more adequate to indicate risk of mortality in this population. Higher BMI protect against mortality when muscle mass was not considered.

Agreement between self-reported and measured weight among older adults in southern Brazil / Concordância entre peso referido e medido entre idosos do sul do Brasil

OBJECTIVE: To evaluate the agreement between self-reported and measured weight among older adults living in the city of Pelotas, southern Brazil. METHODS: Cross-sectional analyses using data from the second follow-up of "COMO VAI?", a longitudinal population-based study that included community-dwelling older adults from the urban region of Pelotas. Weight was self-reported and measured at the same visit. Standing height was estimated based on knee height, which was measured at the first follow-up (2014). Body mass index was categorized as normal (< 24.9 kg/m2), overweight (between 25.0 and 29.9 kg/m2) or obese (≥ 30 kg/m2). Agreement between self-reported and measured weight and between body mass index based on self-reported and measured weight was evaluated using Lin's concordance correlation coefficient and a Bland-Altman plot. The kappa coefficient was used to evaluate the agreement between body mass index according to self-reported and measured weight. RESULTS: Ninety-nine participants were visited at home, where their weight was self-reported and directly measured. The mean difference between measured and self-reported weight was 1.8 kg (95%CI 0.5; 3.1) in men and -0.1 kg (95%CI -1.1; 0.8) in women. Nutritional status according to self-reported and measured weight showed good agreement (81% of the elderly correctly classified themselves) with a kappa of 0.71 in men and 0.68 in women, and a weighted kappa of 0.75 in men and 0.72 in women. CONCLUSIONS: There was good agreement between self-reported and measured weight in women, but strong agreement for nutritional status according to self-reported and measured weight in both sexes, thus self-reported weight can be used to monitor nutritional status in older adults.

OBJETIVO: Avaliar a concordância entre peso autorreferido e medido em idosos de Pelotas, no Sul do Brasil. MÉTODOS: Análise transversal usando dados do segundo acompanhamento do "COMO VAI?", um estudo longitudinal de base populacional que inclui idosos residentes na área urbana de Pelotas, sul do Brasil. Peso autorreferido e medido foram coletados no mesmo momento. Altura em pé foi estimada com base na altura do joelho medida no primeiro seguimento (2014). O índice de massa corporal (IMC) foi classificado como normal (< 24,9 kg/m2), sobrepeso (25,0-29,9 kg/m2) ou obesidade (≥ 30 kg/m2). A concordância entre peso autorreferido e medido e entre o IMC usando peso autorreferido e medido foi avaliada pelo coeficiente de correlação de concordância de Lin e pelo gráfico de Bland-Altman. O coeficiente Kappa foi utilizado para avaliar a concordância entre o estado nutricional do IMC usando peso autorreferido e medido. RESULTADOS: Noventa e nove participantes visitaram domicílios com peso autorreferido e medido. A diferença média entre peso medido e referido foi de 1,8 kg (IC95% 0,5; 3,1) e -0,1 kg (IC95% -1,1; 0,8) para homens e mulheres, respectivamente. O estado nutricional utilizando peso autorreferido e medido mostrou boa concordância (81% dos idosos classificados corretamente), com o kappa de 0,71 e 0,68 e kappa ponderado de 0,75 e 0,72 para homens e mulheres, respectivamente. CONCLUSÕES: Os achados apresentaram boa concordância entre peso autorreferido e medido no sexo feminino, mas forte concordância quanto ao estado nutricional, usando peso autorreferido e medido para ambos os sexos, possibilitando o uso do peso autorreferido para monitorar o estado nutricional em idosos.