Parameter estimation of Poisson generalized linear mixed models based on three different statistical principles : a simulation study

Generalized linear mixed models are flexible tools for modeling non-normal data and are useful for accommodating overdispersion in Poisson regression models with random effects. Their main difficulty resides in the parameter estimation because there is no analytic solution for the maximization of the marginal likelihood. Many methods have been proposed for this purpose and many of them are implemented in software packages. The purpose of this study is to compare the performance of three different statistical principles - marginal likelihood, extended likelihood, Bayesian analysis-via simulation studies. Real data on contact wrestling are used for illustration.

Impacto da estratégia saúde da família com equipe de saúde bucal sobre indicadores de saúde bucal: análise em municípios do nordeste com mais de 100 mil habitantes

This study examined in municipalities of Northeast of Brazil with more than one hundred thousand people who incorporation of Oral Health Teams (OHT) into the Family Health Strategy (FHE) the possible impact on oral health indicators. Sought to answer whether implementation OHT brought the best indicators of health problems and coverage, compared to areas without coverage by the FHE through a community trial in parallel, quasi-randomized. In each of the municipalities surveyed were 20 census tracts, 10 were located in areas covered by oral health teams in the ESF and 10 industries in areas not covered. The final sample consisted of 59.221 individuals. We compared oral health indicators related to health problems, access to services and coverage of oral health actions. The analysis strategy was based on the calculation of prevalence ratios and confidence intervals, adjusted for confounding factors through Poisson regression with robust variance. It also has measured the association between an indicator of social inequality for comparison between areas. The best results are associated with indicators of access and coverage of oral health actions at the expense of the indicators of health problems, suggesting a possible maintenance of a traditional model of practice yet. The results also suggest a possible effect of a specific policy in the area of primary care on inequality in access. From the discussions presented throughout this work, we can see that the impact analysis of public policy, obtained by comparing areas with and without the intervention, not only captures the effect on the target population, but other dimensions of organization service and therefore should be understood as one of the analytical possibilities related to the management

Estimation of genetic parameters for partial egg production periods by means of random regression models

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Variance component estimates applying random regression models for test-day milk yield in Caracu heifers (Bos taurus Artiodactyla, Bovidae)

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Avaliação longitudinal do impacto da saúde bucal na estratégia saúde da família em Natal

As a result of the financial incentive provided by the GM / MS 1.444, since 2000, Brazil has experienced a substantial increase in the number of oral health services at the Family Health Strategy. There is, however, evidence that these teams have produced qualitatively different experiences which do not translate necessarily into improved quality of life and health. Thus, evaluative research of great importance. This study aims to assess the impact of the Family Health Strategy in oral health in a longitudinal perspective natalense the years 2006 and 2009. This is an intervention study whose design is a community trial in parallel, nearly randomized. The sample consisted of census tracts covered by oral health teams in the Family Health and the Traditional Model (Basic Health Units and non-FHS Program of Community Health Agents and areas not covered.) The sample was determined by drawing ten census tracts to form the experimental group and ten other sectors for the control group by pairing intentional based on socio-economic and geographic. To check the net effect of the intervention was performed multivariate analysis by Poisson regression. As a result of cross-sectional analysis of year 2009, it was found that the effects of the ESF in Natal were satisfactory only for the variables of injuries and for other purposes without and with negative impact on stock coverage reclaimers. However, the longitudinal analysis revealed that the ESB / ESF improved their performance in dealing with grievances, access and coverage of the type of actions and this fact is independent of age, sex and social and economic conditions. In other employees' words are related to the presence of the Family Health Strategy in the region. However it does not say that both models under study (the Family Health Strategy Model and Traditional) are different in terms of performance and it is pertinent to reflect on the need for further development of evaluation studies that use other approaches able to clarify the dynamics of the process whose results can come to the knowledge of the actors responsible for leading the ESF and encourage them to incorporate the assessment in their routine

Inatividade física e fatores associados em adultos, São Paulo, Brasil

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Logistic regression analysis of potential prognostic factors for pulmonary thromboembolism

Objective: To identify potential prognostic factors for pulmonary thromboembolism (PTE), establishing a mathematical model to predict the risk for fatal PTE and nonfatal PTE.Method: the reports on 4,813 consecutive autopsies performed from 1979 to 1998 in a Brazilian tertiary referral medical school were reviewed for a retrospective study. From the medical records and autopsy reports of the 512 patients found with macroscopically and/or microscopically,documented PTE, data on demographics, underlying diseases, and probable PTE site of origin were gathered and studied by multiple logistic regression. Thereafter, the jackknife method, a statistical cross-validation technique that uses the original study patients to validate a clinical prediction rule, was performed.Results: the autopsy rate was 50.2%, and PTE prevalence was 10.6%. In 212 cases, PTE was the main cause of death (fatal PTE). The independent variables selected by the regression significance criteria that were more likely to be associated with fatal PTE were age (odds ratio [OR], 1.02; 95% confidence interval [CI], 1.00 to 1.03), trauma (OR, 8.5; 95% CI, 2.20 to 32.81), right-sided cardiac thrombi (OR, 1.96; 95% CI, 1.02 to 3.77), pelvic vein thrombi (OR, 3.46; 95% CI, 1.19 to 10.05); those most likely to be associated with nonfatal PTE were systemic arterial hypertension (OR, 0.51; 95% CI, 0.33 to 0.80), pneumonia (OR, 0.46; 95% CI, 0.30 to 0.71), and sepsis (OR, 0.16; 95% CI, 0.06 to 0.40). The results obtained from the application of the equation in the 512 cases studied using logistic regression analysis suggest the range in which logit p > 0.336 favors the occurrence of fatal PTE, logit p < - 1.142 favors nonfatal PTE, and logit P with intermediate values is not conclusive. The cross-validation prediction misclassification rate was 25.6%, meaning that the prediction equation correctly classified the majority of the cases (74.4%).Conclusions: Although the usefulness of this method in everyday medical practice needs to be confirmed by a prospective study, for the time being our results suggest that concerning prevention, diagnosis, and treatment of PTE, strict attention should be given to those patients presenting the variables that are significant in the logistic regression model.

Utilização de medicamentos e fatores associados: um estudo de base populacional no Município de Campinas, São Paulo, Brasil

Este artigo analisa a prevalência da utilização de medicamentos segundo variáveis demográficas, socioeconômicas e de comportamentos relacionados à saúde, identificando fatores associados ao uso por meio de estudo transversal de base populacional, com 941 pessoas de 18 anos ou mais residentes em Campinas, São Paulo, Brasil. A amostragem foi realizada em múltiplos estágios, estratificada e por conglomerados. Utilizou-se o teste qui-quadrado, foram estimadas as razões de prevalência ajustadas por sexo e idade e os respectivos IC95%. Desenvolveu-se modelo de regressão múltipla de Poisson ficando associados ao uso: sexo feminino, idade de 40 anos e mais, morbidade referida nos últimos 15 dias e número de doenças crônicas. Os medicamentos mais consumidos foram para os sistemas cardiovascular e nervoso, e os fitoterápicos. A prevalência de uso de medicamentos em Campinas encontrou-se inferior à maioria dos estudos. Por meio de inquéritos de saúde locais espera-se conhecer o perfil de uso dos medicamentos pela população e garantir intervenções mais direcionadas para a Política de Assistência Farmacêutica.

Buffalos milk yield analysis using random regression models

Data comprising 1,719 milk yield records from 357 females (predominantly Murrah breed), daughters of 110 sires, with births from 1974 to 2004, obtained from the Programa de Melhoramento Genetic de Bubalinos (PROMEBUL) and from records of EMBRAPA Amazonia Oriental - EAO herd, located in Belem, Para, Brazil, were used to compare random regression models for estimating variance components and predicting breeding values of the sires. The data were analyzed by different models using the Legendre's polynomial functions from second to fourth orders. The random regression models included the effects of herd-year, month of parity date of the control; regression coefficients for age of females (in order to describe the fixed part of the lactation curve) and random regression coefficients related to the direct genetic and permanent environment effects. The comparisons among the models were based on the Akaike Infromation Criterion. The random effects regression model using third order Legendre's polynomials with four classes of the environmental effect were the one that best described the additive genetic variation in milk yield. The heritability estimates varied from 0.08 to 0.40. The genetic correlation between milk yields in younger ages was close to the unit, but in older ages it was low.

Random regression models using Legendre polynomials or linear splines for test-day milk yield of dairy Gyr (Bos indicus) cattle

Studies investigating the use of random regression models for genetic evaluation of milk production in Zebu cattle are scarce. In this study, 59,744 test-day milk yield records from 7,810 first lactations of purebred dairy Gyr (Bos indicus) and crossbred (dairy Gyr × Holstein) cows were used to compare random regression models in which additive genetic and permanent environmental effects were modeled using orthogonal Legendre polynomials or linear spline functions. Residual variances were modeled considering 1, 5, or 10 classes of days in milk. Five classes fitted the changes in residual variances over the lactation adequately and were used for model comparison. The model that fitted linear spline functions with 6 knots provided the lowest sum of residual variances across lactation. On the other hand, according to the deviance information criterion (DIC) and Bayesian information criterion (BIC), a model using third-order and fourth-order Legendre polynomials for additive genetic and permanent environmental effects, respectively, provided the best fit. However, the high rank correlation (0.998) between this model and that applying third-order Legendre polynomials for additive genetic and permanent environmental effects, indicates that, in practice, the same bulls would be selected by both models. The last model, which is less parameterized, is a parsimonious option for fitting dairy Gyr breed test-day milk yield records. © 2013 American Dairy Science Association.

Comparison of random regression models to estimate genetic parameters for milk production in Guzerat (Bos indicus) cows

Random regression models have been widely used to estimate genetic parameters that influence milk production in Bos taurus breeds, and more recently in B. indicus breeds. With the aim of finding appropriate random regression model to analyze milk yield, different parametric functions were compared, applied to 20,524 test-day milk yield records of 2816 first-lactation Guzerat (B. indicus) cows in Brazilian herds. The records were analyzed by random regression models whose random effects were additive genetic, permanent environmental and residual, and whose fixed effects were contemporary group, the covariable cow age at calving (linear and quadratic effects), and the herd lactation curve. The additive genetic and permanent environmental effects were modeled by the Wilmink function, a modified Wilmink function (with the second term divided by 100), a function that combined third-order Legendre polynomials with the last term of the Wilmink function, and the Ali and Schaeffer function. The residual variances were modeled by means of 1, 4, 6, or 10 heterogeneous classes, with the exception of the last term of the Wilmink function, for which there were 1, from 0.20 to 0.33. Genetic correlations between adjacent records were high values (0.83-0.99), but they declined when the interval between the test-day records increased, and were negative between the first and last records. The model employing the Ali and Schaeffer function with six residual variance classes was the most suitable for fitting the data. © FUNPEC-RP.

Genetic evaluation using multi-trait and random regression models in Simmental beef cattle

The Brazilian Association of Simmental and Simbrasil Cattle Farmers provided 29,510 records from 10,659 Simmental beef cattle; these were used to estimate (co)variance components and genetic parameters for weights in the growth trajectory, based on multi-trait (MTM) and random regression models (RRM). The (co)variance components and genetic parameters were estimated by restricted maximum likelihood. In the MTM analysis, the likelihood ratio test was used to determine the significance of random effects included in the model and to define the most appropriate model. All random effects were significant and included in the final model. In the RRM analysis, different adjustments of polynomial orders were compared for 5 different criteria to choose the best fit model. An RRM of third order for the direct additive genetic, direct permanent environmental, maternal additive genetic, and maternal permanent environment effects was sufficient to model variance structures in the growth trajectory of the animals. The (co)variance components were generally similar in MTM and RRM. Direct heritabilities of MTM were slightly lower than RRM and varied from 0.04 to 0.42 and 0.16 to 0.45, respectively. Additive direct correlations were mostly positive and of high magnitude, being highest at closest ages. Considering the results and that pre-adjustment of the weights to standard ages is not required, RRM is recommended for genetic evaluation of Simmental beef cattle in Brazil. ©FUNPEC-RP.

Associações entre nível, oferta de atividade física no trabalho e atividade física de lazer

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Atividade física em gestantes assistidas na atenção primária à saúde

OBJETIVO Analisar o padrão de atividade física de gestantes de baixo risco e os fatores associados. MÉTODOS Estudo transversal com 256 gestantes adultas no segundo trimestre gestacional, sorteadas dentre as assistidas pelas unidades de atenção primária à saúde do município de Botucatu, SP, em 2010. As atividades físicas foram investigadas por meio do “pregnancy physical activity questionnaire”, verificando-se tempo e intensidade de atividades ocupacionais, de deslocamento, domésticas e de lazer, expressos em equivalentes metabólicos dia. As gestantes foram classificadas segundo nível de atividade e em relação a atingir 150 min/semana de atividades físicas de lazer, variáveis dependentes do estudo. A associação entre essas variáveis e as socioeconômicas, características maternas, fatores comportamentais e modelo de atenção da unidade de saúde foi avaliada mediante modelos de regressão de Poisson com variância robusta, adotando-se modelo hierárquico. RESULTADOS A maior parte das gestantes era insuficientemente ativa (77,7%), 12,5% moderadamente ativa e 9,8% vigorosamente ativa. Os maiores gastos diários de energia foram com atividades domésticas, seguidas pelas atividades de locomoção; 10,2% atingiram a recomendação de 150 min semanais de atividades físicas de lazer. Trabalho fora de casa reduziu a chance de atingir essa recomendação (RP = 0,39, IC95% 0,16;0,93). Ter tido pelo menos um parto anterior (RP = 0,87, IC95% 0,77;0,99) e excesso ponderal pré-gestacional (RP = 0,85, IC95% 0,731;0,99) reduziram a chance de ser insuficientemente ativa, enquanto consumir menos alimentos saudáveis teve aumento discreto (RP = 1,18, IC95% 1,02;1,36). CONCLUSÕES Gestantes assistidas na atenção primária à saúde são insuficientemente ativas. Ter tido pelo menos um parto e apresentar sobrepeso pré-gestacional foram identificados como fatores protetores contra tal situação, enquanto consumo menos frequente de alimentos saudáveis foi fator de risco, sugerindo aglomeração de fatores de risco à saúde.

Factors associated with the prevalence of hypertension and control practices among elderly residents of São Paulo city, Brazil

The aim of this study was to analyze the prevalence of hypertension and control practices among the elderly. The survey analyzed data from 872 elderly people in São Paulo, Brazil, through a cluster sampling, stratified according to education and income. A Poisson multiple regression model checked for the existence of factors associated with hypertension. The prevalence of self-reported hypertension among the elderly was 46.9%. Variables associated with hypertension were self-rated health, alcohol consumption, gender, and hospitalization in the last year, regardless of age. The three most common measures taken to control hypertension, but only rarely, are oral medication, routine salt-free diet and physical activity. Lifestyle and socioeconomic status did not affect the practice of control, but knowledge about the importance of physical activity was higher among those older people with higher education and greater income. The research suggests that health policies that focus on primary care to encourage lifestyle changes among the elderly are necessary.