RESUMO
In neuroscience research, longitudinal data are often analysed using analysis of variance (ANOVA) and multivariate analysis of variance (MANOVA) for repeated measures (rmANOVA/rmMANOVA). However, these analyses have special requirements: The variances of the differences between all possible pairs of within-subject conditions (i.e., levels of the independent variable) must be equal. They are also limited to fixed repeated time intervals and are sensitive to missing data. In contrast, other models, such as the generalized estimating equations (GEE) and the generalized linear mixed models (GLMM), suggest another way to think about the data and the studied phenomenon. Instead of forcing the data into the ANOVAs assumptions, it is possible to design a flexible/personalized model according to the nature of the dependent variable. We discuss some advantages of GEE and GLMM as alternatives to rmANOVA and rmMANOVA in neuroscience research, including the possibility of using different distributions for the parameters of the dependent variable, a better approach for different time length points, and better adjustment to missing data. We illustrate these advantages by showing a comparison between rmANOVA and GEE in a real example and providing the data and a tutorial code to reproduce these analyses in R. We conclude that GEE and GLMM may provide more reliable results when compared to rmANOVA and rmMANOVA in neuroscience research, especially in small sample sizes with unbalanced longitudinal designs with or without missing data.
Assuntos
Modelos Estatísticos , Neurociências , Análise de Variância , Projetos de Pesquisa , Modelos Lineares , Estudos LongitudinaisRESUMO
Aedes (Stegomyia) aegypti (L.) (Diptera: Culicidae) is the vector of multiple arboviruses. To evaluate the association between environmental factors and the oviposition activity of Ae. aegypti in Argentina, data on the presence and abundance of eggs were collected using ovitraps, between September of 2018 and May of 2019, in the cities of Villa María, Río Cuarto and Salsipuedes (Córdoba province, Argentina). We analysed the relationships between oviposition and five environmental factors: Temperature, precipitation, vegetation cover, human population density and distance to sites with a potential high density of larval habitats, like cemeteries and trash dumps. Environmental factors' data were collected using satellite image products. The oviposition activity was randomly distributed in three cities. Using generalized linear mixed models, we show that the house where each ovitrap was placed was a source of variability in oviposition, suggesting the relevance of microsite factors and the importance of domestic control actions. Ae. aegypti oviposition was positively correlated with night-time temperature of the previous 3 weeks, and in a context-dependent manner, it was positively correlated with human population density, vegetation cover and precipitation. The consistency and magnitude of these relationships varied between cities, indicating that oviposition is related to a complex system of environmental variables.
Assuntos
Aedes , Animais , Argentina , Feminino , Larva , Mosquitos Vetores , OviposiçãoRESUMO
We introduce a numerically tractable formulation of Bayesian joint models for longitudinal and survival data. The longitudinal process is modeled using generalized linear mixed models, while the survival process is modeled using a parametric general hazard structure. The two processes are linked by sharing fixed and random effects, separating the effects that play a role at the time scale from those that affect the hazard scale. This strategy allows for the inclusion of nonlinear and time-dependent effects while avoiding the need for numerical integration, which facilitates the implementation of the proposed joint model. We explore the use of flexible parametric distributions for modeling the baseline hazard function which can capture the basic shapes of interest in practice. We discuss prior elicitation based on the interpretation of the parameters. We present an extensive simulation study, where we analyze the inferential properties of the proposed models, and illustrate the trade-off between flexibility, sample size, and censoring. We also apply our proposal to two real data applications in order to demonstrate the adaptability of our formulation both in univariate time-to-event data and in a competing risks framework. The methodology is implemented in rstan.
Assuntos
Modelos Estatísticos , Teorema de Bayes , Simulação por Computador , Humanos , Modelos Lineares , Estudos LongitudinaisRESUMO
Quantifying the impacts of disturbances such as oil spills on marine species can be challenging. Natural environmental variability, human responses to the disturbance (e.g., fisheries closures), the complex life histories of the species being monitored, and limited pre-spill data can make detection of effects of oil spills difficult. Using long-term monitoring data from the state of Louisiana (USA), we applied novel spatiotemporal approaches to identify anomalies in species occurrence and catch rates. We included covariates (salinity, temperature, turbidity) to help isolate unusual events. While some species showed evidence of unlikely temporal anomalies in occurrence or catch rates, we found that the majority of the observed anomalies were also before the Deepwater Horizon event. Several species-gear combinations suggested upticks in the spatial variability immediately following the spill, but most species indicated no trend. Across species-gear combinations, there was no clear evidence for synchronous or asynchronous responses in occurrence or catch rates across sites following the spill. Our results are in general agreement to other analyses of monitoring data that detected small impacts, but in contrast to recent results from ecological modeling that showed much larger effects of the oil spill on fish and shellfish.
Assuntos
Pesqueiros/estatística & dados numéricos , Peixes/fisiologia , Poluição por Petróleo/análise , Poluentes Químicos da Água/análise , Animais , Monitoramento Ambiental , Golfo do México , Humanos , Louisiana , Alimentos Marinhos/análise , Análise Espaço-Temporal , Poluição Química da Água/estatística & dados numéricosRESUMO
Se estimaron valores de la talla de madurez gonadal (L50) de la caballa (Scomber japonicus peruanus) para los años 1994 - 2017 en el litoral peruano. El análisis se realizó para el periodo noviembre - marzo donde ocurre la mayor actividad reproductiva. La proporción de hembras maduras en relación a la talla se ajustó mediante un modelo logístico binomial, para ello se utilizaron modelos lineales generalizados con efectos mixtos (MLGM) que asumieron efectos aleatorios asociados a los periodos. El ajuste con el MLGM, mediante sus efectos fjos estimó una L50 de 24.8 cm de longitud a la horquilla (LH) (24.69 cm - 24.9 cm), mientras que con cambios entre periodos, mediante sus efectos aleatorios, los valores oscilaron entre 20.4 cm y 27.0 cm de LH en promedio. Se observó que un aumento en la temperatura causó la disminución de la L50 en los periodos 1997 - 1999 y 2015 - 2016, los cuales coinciden con la presencia de eventos El Niño. Una correlación positiva fue encontrada entre las L50 y la biomasa desovante, estos cambios en las L50 podrían indicar un efecto de denso-dependencia. No se mostró una tendencia en la serie de L50, por tanto no habría un efecto de la presión de pesca, traducida como la tasa instantánea de mortalidad por pesca (F) sobre la L50. Sin embargo, creemos que estas disminuciones de la L50, en relación a una baja densidad poblacional y principalmente a un incremento en la F deben darse como un efecto a largo plazo y no en eventos puntuales como se reflejaron en nuestros resultados.
The values of size-at-gonad maturity (L50) of chub mackerel (Scomber japonicus peruanus) were estimated for the years 1994 - 2017 in the Peruvian littoral. The analysis was performed for the period November - March where the highest reproductive activity occurs. The proportion of mature females was ftted using a logistic binary model, through generalized linear mixed models (MLGM), which assumes random effects associated with the periods. The GLMM, with fixed effects estimated a L50 of 24.8 cm of fork length (FL) (24.69 cm - 24.9 cm), while with changes between periods, with random effects, L50 ranged between 20.4 cm and 27.0 cm FL. It was observed that an increase in temperature caused the decrease of the L50 in the years 1997 - 1999 and 2015 - 2016, which coincides with El Niño events. A positive correlation was found between the L50 and spawning biomass, so changes in L50 may be a density-dependent effect. There was no trend in the L50 series, so there would be no effect of the fishing mortality (F) on the L50. However, we think that the decrease in the L50 in relation to a low population density and mainly an increase in F should be given as a long-term effect and not in isolated events as our results showed.
RESUMO
OBJECTIVES: We explore demographic, temporal and geographic patterns of 256,588 road traffic fatalities from 1998 to 2013 in Mexico, in context of UN´s decade of action for road safety 2010-2020 (DARS). METHODS: Combined traffic mortality data and population counts were analyzed using mixed-effects logistic regression, distinguishing sex-age groups, vulnerable and protected road users, and municipal size. RESULTS: Rapid growth from 1998 to 2008 in traffic mortality rates has been reversed since 2009. Most deaths averted are among young male protected road users (reduction of 0.95 fatalities per 100,000 per year in males 12-49). In spite of a steady decrease over the full study period, mortality rates remain high in vulnerable road users over 50, with a high mortality rate of 26 per 100,000 males over 75 years in 2013. CONCLUSIONS: Progress on the reduction of deaths advances in Mexico, in line with DARS targets. National road safety efforts require strengthening. Initiatives should target vulnerable road users, specifically adults >50 years in urban areas. Strengthening of drink driving programs aimed at young drivers/occupants is promising.
Assuntos
Acidentes de Trânsito/mortalidade , Acidentes de Trânsito/tendências , Condução de Veículo , Segurança , Adolescente , Adulto , Idoso , Criança , Pré-Escolar , Bases de Dados Factuais , Feminino , Humanos , Lactente , Modelos Logísticos , Masculino , México/epidemiologia , Pessoa de Meia-Idade , Adulto JovemRESUMO
Fishing pressure has increased the extinction risk of many elasmobranch (shark and ray) species. Although many countries have established no-take marine reserves, a paucity of monitoring data means it is still unclear if reserves are effectively protecting these species. We examined data collected by a small group of divers over the past 21 years at one of the world's oldest marine protected areas (MPAs), Cocos Island National Park, Costa Rica. We used mixed effects models to determine trends in relative abundance, or probability of occurrence, of 12 monitored elasmobranch species while accounting for variation among observers and from abiotic factors. Eight of 12 species declined significantly over the past 2 decades. We documented decreases in relative abundance for 6 species, including the iconic scalloped hammerhead shark (Sphyrna lewini) (-45%), whitetip reef shark (Triaenodon obesus) (-77%), mobula ray (Mobula spp.) (-78%), and manta ray (Manta birostris) (-89%), and decreases in the probability of occurrence for 2 other species. Several of these species have small home ranges and should be better protected by an MPA, which underscores the notion that declines of marine megafauna will continue unabated in MPAs unless there is adequate enforcement effort to control fishing. In addition, probability of occurrence at Cocos Island of tiger (Galeocerdo cuvier), Galapagos (Carcharhinus galapagensis), blacktip (Carcharhinus limbatus), and whale (Rhincodon typus) sharks increased significantly. The effectiveness of MPAs cannot be evaluated by examining single species because population responses can vary depending on life history traits and vulnerability to fishing pressure.
Modificación del Ensamble de Comunidades de Elasmobranquios en la Isla de Cocos, un Área Marina Protegida Aislada Resumen La presión de la pesca ha incrementado el riesgo de extinción de muchas especies de elasmobranquios (tiburones y rayas). Aunque muchos países han establecido áreas de no-pesca, una escasez de datos de monitoreo implica que todavía no está claro si estas reservas están protegiendo efectivamente a estas especies. Examinamos datos colectados por un pequeño grupo de buzos a lo largo de los últimos 21 años en una de las áreas marinas protegidas (AMP) más vieja del mundo: Parque Nacional Isla de Cocos, Costa Rica. Usamos modelos de efectos mixtos para determinar tendencias en la abundancia relativa, o probabilidad de caso, de doce especies monitoreadas de elasmobranquios compensando la variación entre observadores y de factores abióticos. Ocho de las doce especies declinaron significativamente a lo largo de las últimas dos décadas. Documentamos disminuciones en la abundancia relativa de seis especies, incluidos el tiburón martillo (Sphyrna lewini) (−45%), el tiburón de arrecife de punta blanca (Triaenodon obesus) (−77%), la manta (Mobula spp.) (−78%) y la mantarraya (Manta birostris) (−89%); así como disminuciones en la probabilidad de caso de otras dos especies. Muchas de estas especies tienen extensiones de hábitat pequeñas y deberían estar mejor protegidas por una AMP, lo que enfatiza la noción de que las declinaciones de megafauna marina continuarán sin cesar en las AMP a menos de que exista un esfuerzo adecuado de control de pesca. Además de esto, la probabilidad de ocurrencia en la Isla de Cocos de de los tiburones tigre (Galeocerdo cuvier), de las Galápagos (Carcharhinus galapagensis), de punta negra (Carcharhinus limbatus) y ballena (Rhincodon typus) incrementó significativamente. La efectividad de las AMP no puede ser evaluada examinando a una sola especie porque las respuestas poblacionales pueden variar dependiendo de las características de la historia de vida y de la vulnerabilidad a la presión de la pesca.
Assuntos
Biodiversidade , Conservação dos Recursos Naturais , Tubarões/fisiologia , Rajidae/fisiologia , Animais , Costa Rica , Modelos Biológicos , Oceano Pacífico , Dinâmica PopulacionalRESUMO
Two main classes of methodology have been developed for addressing the analytical intractability of generalized linear mixed models: likelihood-based methods and Bayesian methods. Likelihood-based methods such as the penalized quasi-likelihood approach have been shown to produce biased estimates especially for binary clustered data with small clusters sizes. More recent methods using adaptive Gaussian quadrature perform well but can be overwhelmed by problems with large numbers of random effects, and efficient algorithms to better handle these situations have not yet been integrated in standard statistical packages. Bayesian methods, although they have good frequentist properties when the model is correct, are known to be computationally intensive and also require specialized code, limiting their use in practice. In this article, we introduce a modification of the hybrid approach of Capanu and Begg, 2011, Biometrics 67, 371-380, as a bridge between the likelihood-based and Bayesian approaches by employing Bayesian estimation for the variance components followed by Laplacian estimation for the regression coefficients. We investigate its performance as well as that of several likelihood-based methods in the setting of generalized linear mixed models with binary outcomes. We apply the methods to three datasets and conduct simulations to illustrate their properties. Simulation results indicate that for moderate to large numbers of observations per random effect, adaptive Gaussian quadrature and the Laplacian approximation are very accurate, with adaptive Gaussian quadrature preferable as the number of observations per random effect increases. The hybrid approach is overall similar to the Laplace method, and it can be superior for data with very sparse random effects.